{"id":15654057,"url":"https://github.com/vishal-v/stackgan","last_synced_at":"2025-06-26T20:10:52.416Z","repository":{"id":104746825,"uuid":"203850629","full_name":"Vishal-V/StackGAN","owner":"Vishal-V","description":"TensorFlow implementation of \"Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks\" by Han Zhang, et al.","archived":false,"fork":false,"pushed_at":"2020-04-01T08:55:26.000Z","size":207,"stargazers_count":34,"open_issues_count":5,"forks_count":9,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-04-30T22:33:12.496Z","etag":null,"topics":["conditioning-augmentation","cub-200","gans","generative-adversarial-network","keras","stack-gan","tensorflow-2"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/1612.03242","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Vishal-V.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-08-22T18:25:20.000Z","updated_at":"2024-06-21T08:33:16.000Z","dependencies_parsed_at":null,"dependency_job_id":"5c1bbb3d-8b73-40ab-853e-f3c3534695b1","html_url":"https://github.com/Vishal-V/StackGAN","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Vishal-V/StackGAN","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vishal-V%2FStackGAN","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vishal-V%2FStackGAN/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vishal-V%2FStackGAN/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vishal-V%2FStackGAN/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Vishal-V","download_url":"https://codeload.github.com/Vishal-V/StackGAN/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vishal-V%2FStackGAN/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":262137149,"owners_count":23264675,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["conditioning-augmentation","cub-200","gans","generative-adversarial-network","keras","stack-gan","tensorflow-2"],"created_at":"2024-10-03T12:49:20.960Z","updated_at":"2025-06-26T20:10:52.392Z","avatar_url":"https://github.com/Vishal-V.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# StackGAN\n### Text to Photo-Realistic Image Synthesis\n---\n#### Dependencies\n```\ntensorflow==2.1.0\nnumpy==1.16.4\nabsl_py==0.7.0\nmatplotlib==2.2.3\npandas==0.23.4\nPillow==6.1.0\n```\n#### Downloads\n- To download all the dependencies, simply execute \n```\npip install -r requirements.txt\n```\n- To download the CUB 200 dataset, simply execute the `data_download.py` file\n```\npython data_download.py\n```\n- Download the Char-RNN-CNN embeddings from this link: [**download link**](https://drive.google.com/file/d/0B3y_msrWZaXLT1BZdVdycDY5TEE) and unzip it in place. \n```\nunzip birds.zip\n```\n#### Training\n- The `model.py` file contains the bare minimum code to run the stage 1 and stage 2 architecture. It automatically stores the weights after the specified/default number of epochs have completed. Note that the weights will be stored at the same directory level as `model.py`.\n```\npython model.py\n```\n#### Architecture\n\u003cimg src=\"./assets/stackgan_framework.jpg\" width=\"850px\" height=\"370px\"/\u003e\n\n- Stage 1\n\t- Text Encoder Network\n\t\t- Text description to a 1024 dimensional text embedding\n\t\t- Learning Deep Representations of Fine-Grained Visual Descriptions [Arxiv Link](https://arxiv.org/abs/1605.05395)\n\t- Conditioning Augmentation Network\n\t\t- Adds randomness to the network\n\t\t- Produces more image-text pairs\n\t- Generator Network\n\t- Discriminator Network\n\t- Embedding Compressor Network\n\t- Outputs a 64x64 image\n#\n- Stage 2\n\t- Text Encoder Network\n\t- Conditioning Augmentation Network\n\t- Generator Network\n\t- Discriminator Network\n\t- Embedding Compressor Network\n\t- Outputs a 256x256 image\n---\n#### Reference Papers\n1. **StackGAN: Text to photo-realistic image synthesis** [[Arxiv Link](https://arxiv.org/pdf/1612.03242.pdf)]\n2. **Improved Techniques for Training GANs** [[Arxiv Link](https://arxiv.org/pdf/1606.03498.pdf)]\n3. **Generative Adversarial Text to Image Synthesis** [[Arxiv Link](https://arxiv.org/pdf/1605.05396.pdf)]\n4. **Learning Deep Representations of Fine-Grained Visual Descriptions** [[Arxiv Link](https://arxiv.org/abs/1605.05395)]\n---\n#### Note\nThis is the code I have submitted to TensorFlow for Google Summer of Code. Hence the attributions and the License is for \"TensorFlow Authors\" and not \"Vishal V\". This code is under the MIT License.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvishal-v%2Fstackgan","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvishal-v%2Fstackgan","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvishal-v%2Fstackgan/lists"}